基于高光谱数据的三文鱼掺水无损鉴别

摘要：针对三文鱼掺水鉴别，提出了一种基于高光谱图像的三文鱼掺水无损鉴别方法。利用系统获取了可见光和近红外 波段（390-1050nm）的三文鱼片高光谱图像。对原始高光谱数据进行主成分分析（PCA），根据图像质量和 PCA 参数， 选取第二主成分（PC2）图像作为特征图像，提取特征图像加权系数局部极值对应的波长作为特征波长，分别为 454.9、 512.3、569.1nm。在此基础上，分别以颜色结合特征波长光谱、纹理结合特征波长光谱、颜色-纹理结合特征波长光谱为 输入，建立了基于自组织特征映射（SOM）网络的三文鱼掺假识别建模。分析了模型相邻神经元距离和特征权重，实现 了识别结果的可视化。结果表明，以纹理-颜色结合特征波长光谱的融合特征为输入的基于 SOM 网络的模型性能最优， 识别准确率高达96.7%。

期刊：Journal of Food Quality

DOI：https://DOI.org/10.1155/2018/1809297

设备：GaiaSorter

Nondestructive Identification of Salmon Adulteration with Water Based on Hyperspectral Data

Abstract：For the identiŒcation of salmon adulteration with water injection, a nondestructive identiŒcation method based on hyperspectral images was proposed. …ehyperspectral images of salmon Œllets in visible and near-infrared ranges (390–1050nm) were obtained with a system. eoriginal hyperspectral data were processed through the principal-component analysis (PCA). According to the image quality and PCA parameters, a second principal-component (PC2) image was selected as the feature image, and the wavelengths corresponding to the local extremum values of feature image weighting coe™cients were extracted as feature wavelengths, which were 454.9, 512.3, and 569.1nm. Onthis basis, the color combined with spectra at feature wavelengths, texture combined with spectra at feature wavelengths, and color-texture combined with spectra at feature wavelengths were in dependently set as the input, for the modeling of salmon adulteration identiŒcation based on the self-organizing feature map (SOM) network.e distances between neighboring neurons and feature weights of the models were analyzed to realize the visualization of identiŒcation results.e results showed that the SOM-based model, with texture-color combined with fusion features of spectra at feature wavelengths as the input, was evaluated to possess the best performance and identiŒcation accuracy is as high as 96.7%.